Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
Phylogenetic diversity and conservation
Dan Faith
The Australian Museum
Applied ecology and human dimensions in biological conservation
Biota Program/ FAPESP
Nov. 9-10, 2009
BioGENESIS
Providing an
evolutionary
framework for
biodiversity
science
A global Biodiversity Observation Network:
GEO BON
Adopting systematic conservation planning
can mean a region shifts to a better trajectory
Faith, DP & Ferrier S (2005)
Good news and bad news for the 2010 biodiversity target. Science Online
Ecosystem services
• Seems to dominate ideas on post-2010 targets
• Is this all that matters?
• Sometimes species counts are considered, as capturing ethical values, existence values etc.– But this misses the point.......
Evosystem services
• Consider the “evolutionary system” and its services
• This better captures “option values” or future, often
unanticipated, benefits/uses
• Species counts may measure this, but another good
way is to measure the evolutionary history that is
preserved
– Indicates known products and scope for finding new
products not yet known
– Also indicates capacity for producing new diversity,
including in response to global changes
Adopting systematic conservation planning
can mean a region shifts to a better trajectory
Faith, DP & Ferrier S (2005)
Good news and bad news for the 2010 biodiversity target. Science Online
The
phylogenetic
diversity
measure,
“PD”
can
contribute
to this
strategy
PD – phylogenetic diversityPD of a set of taxa = length of spanning path of the set on the phylogeny
(how much of the tree travelled over if connect up those taxa on the tree)
Tick marks along the branches = character changes
PD measures “feature diversity”
scenario B represents more feature diversity
Faith 1992
Forest et al. (2007)
Nature
blue = genera in the Cape having species
of medicinal or economic importance(as recorded in Survey of Economic Plants for Arid and Semi-Arid Lands)
• small loss of PD or
evolutionary potential for
given species loss
• large loss of PD or
evolutionary potential
red = surviving evolutionary potential
Will the impacts of climate change on PD be large or small?
Yesson, C. and A. Culham. 2006.
PD and probabilities of extinction
Probabilistic PD
– Red numbers
are estimated
probabilities of
extinction
Can estimate “expected phylogenetic diversity” or do
“phylogenetic risk analysis”
.4
.95
.5
.5
Faith DP (2008) Threatened species and the preservation of phylogenetic diversity (PD):
assessments based on extinction probabilities and risk analysis. Conservation Biology
www.edgeofexistence.org
Mooers AØ, Faith DP, Maddison WP (2008) Converting Endangered Species Categories
to Probabilities of Extinction for Phylogenetic Conservation Prioritization. PLoS ONE
Loss of the world‟s corals• Number of threatened species –
• “the proportion of corals (57.8%) exceeds that of all
terrestrial animal groups assessed to date..” • Carpenter et al (2008) Science
many examples where entire clades (existing families and genera) fall into
IUCN threatened classes
- apply phylogenetic risk analysis to a “supertree” for corals.
Reducing the slide from near-threatened to
threatened status
Change from near-threatened to threatened
Bio
div
ers
ity =
evolu
tionary
his
tory
Shift to a better
trajectory of change
Perhaps risk analysis can help identify key places to invest in
coral conservation, to avoid worst-case PD losses
GBIF (Global Biodiversity Information Facility) and other data
A global Biodiversity Observation Network:
how do we monitor genetic/phylogenetic diversity?
GEO BON
Remotely mapped
climate, terrain &
substrate data
Phylogenetic
patterns
Time series of
remotely sensed
land
condition/cover
observations
THE LENS
Biodiversity distribution modelling –
Estimation of status of biodiversity
for different places at different times
Assessment of predicted change in loss
rates & achievement of 2010 target
Conservation &
land-use planning
decisions/scenarios
GEO BON
Phylogenetic ecology
based on PD and feature diversity
We can take any conventional species-level index
and re-express as a PD-based measure
• Richness
• PD-Complementarity (gains & losses)
• PD-Endemism (e.g. Faith et al 2004)
• PD-Dissimilarity...
Orange protected area, with protected PD in green
For this region, PD complementarity in red
For this region, PD complementarity in red
Orange protected area, with protected PD in green
DEH Technical Workshop, Nov.
2006
PD and the Cape hotspot: species counting highlights the
western portion but PD highlights the eastern portion
Forest et al
Nature 2007
the PD that you could gain
does
not
match
the PD that you do gain
PD conservation planning
example
• Sydney water supply catchment region
• Threats from mining, new dams
• Baker et al studies, using DNA
barcoding type methods, reveal
lots of cryptic variation over several
freshwater invertebrate groups
+ lots of “phylogeographic” structure
Sydney
L Burragorang
20 km
EDCBA
Key
AUSTRALIA
Refer Detail
Baker et al
Lineage distributions
localities already
impacted or
under threat from
mining and/or
future dams
Euastacus
Upper Georges
Upper Nepean
Upper Georges River‟s unique contribution to
phylogenetic diversity is shown in red
Consider consequences of loss of diversity,
from the human impacts on Upper Nepean River
Faith, D. P and A. Baker, 2006. Phylogenetic diversity (PD) and biodiversity
conservation: some bioinformatics challenges Evolutionary Bioinformatics
see also Faith, D. P. 2008. Phylogenetic diversity and conservation. In (eds: SP Carroll
and C Fox) Conservation Biology: Evolution in Action. Oxford University Press.
For conservation planning
congruence in PD complementarity values
(the red bits)
gives confidence that these values
may reflect more general values
New technologies such as DNA barcoding can help get
more data for more taxonomic groups….
DNA barcoding
• The use of a small, standardised, portion
of DNA sequence for the purpose of
species identification and discovery
• We can use PD calculations – e.g. PD complementarity
(marginal gains and losses)
• Not only do we by-pass species, but we may gain a
boost in prediction of general biodiversity patterns
See e.g. Faith and Williams 2005; Faith and Baker 2006
example
Barcoding and phylogeny estimates
Numbers are
sites
Arctic Collembola PD and the 2010
Biodiversity Target
0 1 2 3 4
number of sites lost
am
ou
nt
of
PD
co
nserv
ed
Prospects: a toolbox for application to “phylogenies” from large scale DNA
barcoding programs
Faith, DP (2008) Phylogenetic diversity and conservation. In (eds: SP Carroll and
C Fox) Conservation Biology: Evolution in Action. Oxford University Press.
PD-complementarity values
often missing information for many sites...
Lozupone and Knight 2005, Ferrier et al 2007
extend PD to produce dissimilarities between sites.
These are linked to environmental variables
Have used PD to quantify biodiversity -
“The explosion of 16S rRNA gene sequence
data in the public databases, in conjunction
with new high-throughput sequencing
technologies such as pyrosequencing...
allows us to address a vast range of
fundamental questions about microbial
communities on an unprecedented scale.”
“soil, which is often described as one of the
most diverse types of microbial communities
based on species-based diversity measures,
has below average PD in the bacteria”
(Lozupone et al., review, 2008)
Big
trees,
Few
taxa,
Many
samples
Microbial ecology
two sample sites j and kj and k dissimilar if lots of red and blue
Big
trees,
Few
taxa,
Many
samples
Environmental gradient
PD-dissimilarities reflect distances along gradients
Microbial ecology
Lozupone and Knight „s “phylogenetic beta
diversity” for global bacteria samples
• Use
phylogenetic
dissimilarity
(“UniFrac”)
among samples
• Discover that the
major environmental
determinant of
microbial community
composition is salinity
Phylogeny helps find important gradients, because even
deeper branches have unimodal response to gradients
Faith, D. P., C. A. Lozupone, D. Nipperess, R. Knight (2009)
A general model linking evolutionary features and environmental
gradients supports broad applications of microbial ecology‟s phylogenetic
beta diversity framework. International Journal of Molecular Science
House dust
communities
• Use
phylogenetic
dissimilarity
among samples
• Discover that the
major environmental
determinant of
microbial community
composition is salinity
Lozupone and Knight „s “phylogenetic beta
diversity” for global bacteria samples
ED (environmental diversity) method “counts-up” species under
evolutionary model of unimodal response to gradients.
ED estimates the biodiversity represented by any set of protected areas
phylogenetic information can be
integrated into a “lens” for interpreting
remotely sensed changes in land
condition – useful for GEO BON....
Remotely mapped
climate, terrain &
substrate data
Phylogenetic
patterns
Time series of
remotely sensed
land
condition/cover
observations
THE LENS
Biodiversity distribution modelling –
using PD-dissimilarities
Estimation of status of biodiversity
for different places at different times
Assessment of predicted change in loss
rates & achievement of 2010 target
Conservation &
land-use planning
decisions/scenarios
GEO BON
A global Biodiversity Observation Network:
how do we monitor genetic/phylogenetic diversity?
GEO BON
Extending the lens approach to monitor impacts at
new sites (no need to have all sites in the space)
Phylogeny helps find important gradients, because even
deeper branches have unimodal response to gradients
Faith, D. P., C. A. Lozupone, D. Nipperess, R. Knight
A general model linking evolutionary features and environmental
gradients supports broad applications of microbial ecology‟s phylogenetic
beta diversity framework. International Journal of Molecular Science
House dust
communities
two dimensional gradient space with reference sample sites = solid and hollow dots
Test site (red dot, green dot) positioned in the space based on environmental values.
Phylogenetic monitoring to detect impacts
We can use the special regression (GDM) to predict PD-
dissimilarities from environmental distances
Monitoring for impacts on biodiversity -
placing new sites into the environmental space
• Convex
shapes are
two different
deep
branches,
solid dot =
branch
present
in site
Monitoring for impacts on biodiversity -
placing new sites into the environmental space
• on the left, absence of
that lineage in the red-
dot site is improbable
• - all reference sites in
the convex hull had
the lineage.
• on the right, absence
of that lineage in the
green-dot site is
probable
• - the lineage is absent
in some reference
sites within the convex
hull
Remotely mapped
climate, terrain &
substrate data
Phylogenetic
patterns
Further investigation of
possible impacts
GEO BON
Placement of a new
site in the model plus
observations on its
current community
Estimation of status of biodiversity
at the new site
THE LENSBiodiversity distribution modelling –
Dissimilarities between sites based on
phylogeny , not species
Conclusions
• “evosystem services” helps better capture “option
values” = future, often unanticipated, benefits/uses
• a good measure is the evolutionary history (PD)
that is preserved
• we can make effective use of phylogeny and new
and old survey data to create models of
biodiversity that act as a “lens” to interpret time
series of changes in land condition
• the same lens models may also allow impact
detection at new sites
• new technologies – DNA barcoding and advanced
sequencing can provide the needed data